MaltSci 麦伴科研

Appearance

Curated Papers > High-impact authoritative literature on "AI Drug Discovery"

Key challenges for delivering clinical impact with artificial intelligence. ​

Literature Information ​

DOI10.1186/s12916-019-1426-2
PMID31665002
JournalBMC medicine
Impact Factor8.3
JCR QuartileQ1
Publication Year2019
Times Cited681
KeywordsAlgorithms, Artificial intelligence, Evaluation, Machine learning, Regulation
Literature TypeJournal Article, Research Support, Non-U.S. Gov't
ISSN1741-7015
Pages195
Issue17(1)
AuthorsChristopher J Kelly, Alan Karthikesalingam, Mustafa Suleyman, Greg Corrado, Dominic King

TL;DR ​

This article highlights the rapid advancement of artificial intelligence (AI) in healthcare, while addressing significant challenges in translating these technologies from research to clinical practice, including the need for robust clinical evaluation and regulation. The authors emphasize that overcoming issues such as algorithmic bias, generalizability, and interpretability is crucial for ensuring that AI systems can safely and effectively improve patient outcomes.

Search for more papers on MaltSci.com ​

Algorithms · Artificial intelligence · Evaluation · Machine learning · Regulation

Abstract ​

BACKGROUND Artificial intelligence (AI) research in healthcare is accelerating rapidly, with potential applications being demonstrated across various domains of medicine. However, there are currently limited examples of such techniques being successfully deployed into clinical practice. This article explores the main challenges and limitations of AI in healthcare, and considers the steps required to translate these potentially transformative technologies from research to clinical practice.

MAIN BODY Key challenges for the translation of AI systems in healthcare include those intrinsic to the science of machine learning, logistical difficulties in implementation, and consideration of the barriers to adoption as well as of the necessary sociocultural or pathway changes. Robust peer-reviewed clinical evaluation as part of randomised controlled trials should be viewed as the gold standard for evidence generation, but conducting these in practice may not always be appropriate or feasible. Performance metrics should aim to capture real clinical applicability and be understandable to intended users. Regulation that balances the pace of innovation with the potential for harm, alongside thoughtful post-market surveillance, is required to ensure that patients are not exposed to dangerous interventions nor deprived of access to beneficial innovations. Mechanisms to enable direct comparisons of AI systems must be developed, including the use of independent, local and representative test sets. Developers of AI algorithms must be vigilant to potential dangers, including dataset shift, accidental fitting of confounders, unintended discriminatory bias, the challenges of generalisation to new populations, and the unintended negative consequences of new algorithms on health outcomes.

CONCLUSION The safe and timely translation of AI research into clinically validated and appropriately regulated systems that can benefit everyone is challenging. Robust clinical evaluation, using metrics that are intuitive to clinicians and ideally go beyond measures of technical accuracy to include quality of care and patient outcomes, is essential. Further work is required (1) to identify themes of algorithmic bias and unfairness while developing mitigations to address these, (2) to reduce brittleness and improve generalisability, and (3) to develop methods for improved interpretability of machine learning predictions. If these goals can be achieved, the benefits for patients are likely to be transformational.

MaltSci.com AI Research Service ​

Intelligent ReadingAnswer any question about the paper and explain complex charts and formulas
Locate StatementsFind traces of a specific claim within the paper
Add to KBasePerform data extraction, report drafting, and advanced knowledge mining

Primary Questions Addressed ​

  1. What specific strategies can be employed to enhance the generalizability of AI algorithms in diverse patient populations?
  2. How can healthcare organizations effectively address sociocultural barriers to the adoption of AI technologies in clinical settings?
  3. What role do regulatory frameworks play in balancing innovation with patient safety in the deployment of AI in healthcare?
  4. In what ways can performance metrics for AI systems be designed to reflect real-world clinical applicability beyond technical accuracy?
  5. How can developers of AI algorithms proactively identify and mitigate potential biases in their datasets to ensure equitable health outcomes?

Key Findings ​

Background and Objectives ​

The article by Kelly et al. (2019) discusses the rapid advancement of artificial intelligence (AI) in healthcare, highlighting its potential to transform various medical domains. However, the authors note a significant gap between AI research and its clinical application. The objective of the paper is to identify the key challenges and limitations that hinder the effective translation of AI technologies from research into clinical practice.

Main Methods/Materials/Experimental Design ​

The authors conducted a comprehensive review of the literature on AI applications in healthcare, focusing on the challenges faced in translating these technologies into clinical settings. The discussion is structured around several critical themes, including:

  1. Clinical Evaluation: Emphasizing the need for robust peer-reviewed clinical trials and appropriate performance metrics that reflect real-world applicability.
  2. Regulatory Frameworks: Addressing the necessity for regulations that balance innovation with patient safety.
  3. Bias and Generalization: Exploring issues related to algorithmic bias and the generalizability of AI systems across diverse patient populations.
Mermaid diagram

Key Results and Findings ​

  • Limited Clinical Deployment: Despite numerous studies demonstrating AI's capabilities, few algorithms have been successfully implemented in clinical practice.
  • Need for Prospective Studies: Most AI research relies on retrospective data, which may not accurately predict real-world performance.
  • Performance Metrics: Current metrics often fail to capture the clinical applicability of AI systems, emphasizing the need for measures that consider patient outcomes and care quality.
  • Regulatory Challenges: Existing regulatory frameworks are often ill-equipped to handle the rapid evolution of AI technologies.

Main Conclusions/Significance/Innovation ​

The authors conclude that for AI to have a meaningful impact in healthcare, there must be a concerted effort to address the challenges of clinical evaluation, regulation, and algorithmic bias. The paper highlights the importance of developing interpretable AI systems that clinicians can trust and understand. If these challenges are addressed, AI could significantly enhance patient care, reduce clinical errors, and improve healthcare efficiency.

Research Limitations and Future Directions ​

  • Limitations: The review primarily discusses challenges without providing empirical data or case studies that illustrate successful AI implementations.
  • Future Directions: The authors suggest several areas for future research, including:
    • Developing methods to mitigate algorithmic bias and improve generalizability.
    • Establishing regulatory frameworks that can adapt to ongoing AI innovations.
    • Enhancing the interpretability of AI systems to facilitate clinician trust and understanding.

In summary, while the promise of AI in healthcare is substantial, realizing its full potential requires overcoming significant hurdles related to clinical validation, regulation, and societal impact.

References ​

  1. Identifying facial phenotypes of genetic disorders using deep learning. - Yaron Gurovich;Yair Hanani;Omri Bar;Guy Nadav;Nicole Fleischer;Dekel Gelbman;Lina Basel-Salmon;Peter M Krawitz;Susanne B Kamphausen;Martin Zenker;Lynne M Bird;Karen W Gripp - Nature medicine (2019)
  2. Caveats for the use of operational electronic health record data in comparative effectiveness research. - William R Hersh;Mark G Weiner;Peter J Embi;Judith R Logan;Philip R O Payne;Elmer V Bernstam;Harold P Lehmann;George Hripcsak;Timothy H Hartzog;James J Cimino;Joel H Saltz - Medical care (2013)
  3. Reporting of artificial intelligence prediction models. - Gary S Collins;Karel G M Moons - Lancet (London, England) (2019)
  4. Artificial intelligence using deep learning to screen for referable and vision-threatening diabetic retinopathy in Africa: a clinical validation study. - Valentina Bellemo;Zhan W Lim;Gilbert Lim;Quang D Nguyen;Yuchen Xie;Michelle Y T Yip;Haslina Hamzah;Jinyi Ho;Xin Q Lee;Wynne Hsu;Mong L Lee;Lillian Musonda;Manju Chandran;Grace Chipalo-Mutati;Mulenga Muma;Gavin S W Tan;Sobha Sivaprasad;Geeta Menon;Tien Y Wong;Daniel S W Ting - The Lancet. Digital health (2019)
  5. The triple aim: care, health, and cost. - Donald M Berwick;Thomas W Nolan;John Whittington - Health affairs (Project Hope) (2008)
  6. Design Characteristics of Studies Reporting the Performance of Artificial Intelligence Algorithms for Diagnostic Analysis of Medical Images: Results from Recently Published Curated Papers. - Dong Wook Kim;Hye Young Jang;Kyung Won Kim;Youngbin Shin;Seong Ho Park - Korean journal of radiology (2019)
  7. Adversarial attacks on medical machine learning. - Samuel G Finlayson;John D Bowers;Joichi Ito;Jonathan L Zittrain;Andrew L Beam;Isaac S Kohane - Science (New York, N.Y.) (2019)
  8. Making Machine Learning Models Clinically Useful. - Nigam H Shah;Arnold Milstein;Steven C Bagley PhD - JAMA (2019)
  9. Measurement Is Essential for Improving Diagnosis and Reducing Diagnostic Error: A Report From the Institute of Medicine. - Elizabeth A McGlynn;Kathryn M McDonald;Christine K Cassel - JAMA (2015)
  10. Predicting scheduled hospital attendance with artificial intelligence. - Amy Nelson;Daniel Herron;Geraint Rees;Parashkev Nachev - NPJ digital medicine (2019)

Literatures Citing This Work ​

  1. Clinical-grade Computational Pathology: Alea Iacta Est. - Filippo Fraggetta - Journal of pathology informatics (2019)
  2. Artificial Intelligence in Medicine: Today and Tomorrow. - Giovanni Briganti;Olivier Le Moine - Frontiers in medicine (2020)
  3. Presenting machine learning model information to clinical end users with model facts labels. - Mark P Sendak;Michael Gao;Nathan Brajer;Suresh Balu - NPJ digital medicine (2020)
  4. The challenges of colposcopy for cervical cancer screening in LMICs and solutions by artificial intelligence. - Peng Xue;Man Tat Alexander Ng;Youlin Qiao - BMC medicine (2020)
  5. Age and sex affect deep learning prediction of cardiometabolic risk factors from retinal images. - Nele Gerrits;Bart Elen;Toon Van Craenendonck;Danai Triantafyllidou;Ioannis N Petropoulos;Rayaz A Malik;Patrick De Boever - Scientific reports (2020)
  6. A data-driven framework for selecting and validating digital health metrics: use-case in neurological sensorimotor impairments. - Christoph M Kanzler;Mike D Rinderknecht;Anne Schwarz;Ilse Lamers;Cynthia Gagnon;Jeremia P O Held;Peter Feys;Andreas R Luft;Roger Gassert;Olivier Lambercy - NPJ digital medicine (2020)
  7. Deep Learning for Accurate Diagnosis of Liver Tumor Based on Magnetic Resonance Imaging and Clinical Data. - Shi-Hui Zhen;Ming Cheng;Yu-Bo Tao;Yi-Fan Wang;Sarun Juengpanich;Zhi-Yu Jiang;Yan-Kai Jiang;Yu-Yu Yan;Wei Lu;Jie-Min Lue;Jia-Hong Qian;Zhong-Yu Wu;Ji-Hong Sun;Hai Lin;Xiu-Jun Cai - Frontiers in oncology (2020)
  8. "Yes, but will it work for my patients?" Driving clinically relevant research with benchmark datasets. - Trishan Panch;Tom J Pollard;Heather Mattie;Emily Lindemer;Pearse A Keane;Leo Anthony Celi - NPJ digital medicine (2020)
  9. Finding undiagnosed patients with hepatitis C infection: an application of artificial intelligence to patient claims data. - Orla M Doyle;Nadejda Leavitt;John A Rigg - Scientific reports (2020)
  10. Accuracy and efficiency of an artificial intelligence tool when counting breast mitoses. - Liron Pantanowitz;Douglas Hartman;Yan Qi;Eun Yoon Cho;Beomseok Suh;Kyunghyun Paeng;Rajiv Dhir;Pamela Michelow;Scott Hazelhurst;Sang Yong Song;Soo Youn Cho - Diagnostic pathology (2020)

... (671 more literatures)

© 2025 MaltSci - We reshape scientific research with AI technology